The polymorphism of alkaline phosphatase in the human placenta has been thoroughly documented, and is unique in the number of rare electrophoretic variants, in comparison with 43 other enzyme loci which have been carefully examined. One such variant, the "D-variant", is found in 0.4 percent (Boston) to 1.0 percent (Upsalla) of normal deliveries. This variant appears to be associated with abortion and complications of pregnancy. It is easier to assay than the electrophoretically defined phenotypes, because it is more sensitive to inhibition by L-leucine. It is therefore proposed that a large number of pregnancies be tested for the presence of this variant, to get a more precise idea of its relationship to high-risk pregnancies. Also, the placental electrophoretic phenotypes will be defined for all placentas from pregnancies with a suboptimal outcome. The effort to determine the relationship between these enzyme variants and complications of pregnancy will be complemented by an intensive research effort on the biochemical properties of the phenotypes, including the purification and characterization of individual variant enzymes and their molecular-weight subforms, isolation of "native" inhibitor (presumed to be a peptide) of the enzyme, and measurement of the in vitro synthesis of the enzyme by preparations of cells from placental tissue. Together, these studies should give a much clearer picture of the significance of the disproportionate polymorphic variation at the locus of this enzyme, and may well help to define a significant contributing factor to high risk pregnancies.